The project is to evaluate the effectiveness of microfinance, specifically the large-scale "Million Baht Village Bank" initiative of the Thai government, in promoting economic development, ameliorating poverty, and improving the outcomes of households in rural and semi-urban Thailand. In the context of theories of growth and financial intermediation, we intend to evaluate the program's impacts on: a) consumption, b) input use and investment, c) asset accumulation, d) occupational mobility, e) entrepreneurship, f) children's education, and g) the markets for alternative credit sources. The unique policy (or "natural") experiment is well suited for evaluating theories of credit-constrained household behavior. Specifically, theories have predictions for the types of households likely to be constrained, and the behavior these households should exhibit, but existing evidence of credit constraints is largely non-experimental. The experiment will enable a test of whether households predicted to be constrained indeed respond to an injection of credit as theory would predict. For example, in a buffer stock saving model with indivisible investments, households with low net assets (i.e. constrained households), bad income shocks in a given year, or large investments are ceteris paribus likely to have higher than average consumption growth because of their inability to borrow sufficiently. The theory underscores the importance of credit constraints and investment indivisibilities in consumption and investment dynamics. The study will test whether the strength of credit constraints (i.e. the Lagrange multiplier on the borrowing constraint) falls in relation to the amount of the exogenous credit increase. The importance of growth and financial intermediation on children and families is best illustrated through its effect on poverty rates. Using the Thai growth experience as an example, despite an increase in inequality, the dramatic economic growth in Thailand between 1976-1996 lowered Thai poverty rates, based on the standard poverty line of two dollars a day, from 48 percent to 13 percent (Jeong, 2000).